Subject: Re: twe status queries?
To: Gordon Waidhofer <firstname.lastname@example.org>
From: Jonathan Stone <jonathan@Pescadero.dsg.stanford.edu>
Date: 12/02/2005 11:44:41
>You could be getting burned by a slow drive.
>Successful retries would not get reported as
>a hard error. You can try swapping drives, one
>position at a time, between the two RAID sets
>and see if the problem moves.
Is there a way to get SMART stats, retry counters, or other
drive-health statistics out through the twe?
Some vendors sell drives specifically tuned to give up early on
error-correction (Reed-Solomon? Viterbi? trellis? I dunno, these
days), but if der Mouse already has drives, that might not be an
Gordon, do you know if there are effective ways (a' la SCSI
mode-pages) to tell commodity ATA/SATA drives to give up early and
report railure, rather than resorting to rereads and error-correction